On 10/14/2013 10:12 AM, Anuj Phogat wrote:
> Implement the FS backend for new builtins added by the extension:
> in vec2 gl_SamplePosition
> in int gl_SampleID
> in int gl_NumSamples
> out int gl_SampleMask[]
>
> Signed-off-by: Anuj Phogat <anuj.pho...@gmail.com>
> ---
> src/mesa/drivers/dri/i965/brw_fs.cpp | 109
> +++++++++++++++++++++++++++
> src/mesa/drivers/dri/i965/brw_fs.h | 4 +
> src/mesa/drivers/dri/i965/brw_fs_visitor.cpp | 23 ++++++
> src/mesa/drivers/dri/i965/brw_wm.h | 1 +
> 4 files changed, 137 insertions(+)
>
> diff --git a/src/mesa/drivers/dri/i965/brw_fs.cpp
> b/src/mesa/drivers/dri/i965/brw_fs.cpp
> index e5d6e4b..e4f7745 100644
> --- a/src/mesa/drivers/dri/i965/brw_fs.cpp
> +++ b/src/mesa/drivers/dri/i965/brw_fs.cpp
> @@ -1115,6 +1115,109 @@
> fs_visitor::emit_frontfacing_interpolation(ir_variable *ir)
> return reg;
> }
>
> +void
> +fs_visitor::compute_sample_position(fs_reg dst, fs_reg int_sample_pos)
> +{
> + int num_samples = ctx->DrawBuffer->Visual.samples;
> + assert(num_samples >= 0 && num_samples <= 8);
> +
> + /* From arb_sample_shading specification:
> + * "When rendering to a non-multisample buffer, or if multisample
> + * rasterization is disabled, gl_SamplePosition will always be
> + * (0.5, 0.5).
> + */
> + if (!ctx->Multisample.Enabled || num_samples == 0) {
> + emit(BRW_OPCODE_MOV, dst, fs_reg(0.5f));
> + }
> + else {
> + /* For num_samples = {4, 8} */
> + emit(BRW_OPCODE_MOV, dst, int_sample_pos);
> + emit(BRW_OPCODE_MUL, dst, dst, fs_reg(1 / 16.0f));
> + }
> +}
> +fs_reg *
> +fs_visitor::emit_sampleid_interpolation(ir_variable *ir)
> +{
> + assert(brw->gen >= 6);
> + bool multisampled_fbo = ctx->DrawBuffer->Visual.samples > 1;
> +
> + this->current_annotation = "compute sample id";
> + fs_reg *reg = new(this->mem_ctx) fs_reg(this, ir->type);
> +
> + if (multisampled_fbo && ctx->Multisample.Enabled) {
One more thing. These two blocks have different conditions. For
gl_SamplePosition, num_samples == 1 hits the MSAA case. For
gl_SampleId, num_samples == 1 hits the non-MSAA case.
I don't know which is right, but they should match.
_______________________________________________
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
